OpenSearch

Thread Pool Rejection Storm

critical
Resource ContentionUpdated Feb 18, 2026

When ThreadPool_QueueSize reaches ThreadPool_QueueCapacity, OpenSearch begins rejecting requests, causing search failures and degraded cluster performance. This often cascades as rejected requests retry.

How to detect:

Monitor ThreadPool_RejectedReqs increasing alongside ThreadPool_QueueSize approaching ThreadPool_QueueCapacity. Look for ThreadPool_QueueLatency increases and ThreadPool_ActiveThreads at or near ThreadPool_TotalThreads.

Recommended action:

Identify the saturated thread pool type (search, write, etc.), investigate queries causing thread exhaustion via Top N queries API, consider tuning thread pool sizes, reduce query complexity, implement query throttling, or scale cluster to handle load.